Federal funding for academic research, especially in the life sciences, has risen rapidly over the last decade. The oft-cited justification for this level of public expenditure is that public funding for biomedical research results in creation of knowledge and expansion of the scientific workforce, thereby leading to better medical treatments and preemption of disease. In this research, we aim to evaluate whether the potential scientific workforce benefits of federal R&D funding are realized in practice. Increases in NIH R&D funding, specifically the NIH post-1998 budget doubling, have been associated with increases in postdoctoral appointments and other soft money positions. However, while intuitively an increase in university R&D funding could yield an increase in demand for graduate research assistants, and therefore expand the cohort of admitted students, universities might also simply cut back on institutional funding for those students, with no net impact on production of trained graduates. This project will use survey data collected on a panel of over 200 US universities and an instrumental variables approach to investigate the causal relationship between changes in NIH R&D funding at US universities and the number and diversity of graduate students enrolled and completing their PhDs in the biomedical sciences. We will also investigate how changes in R&D funding affect the number and composition of postdocs appointed in these fields. Measures of diversity will include biomedical scientists' gender, race and ethnicity, and citizenship/immigration status. The project will also explore possible heterogeneity in effects of federal R&D funding across different types of academic institutions.